Explicit and implicit depth-cue integration: Evidence of systematic biases with real objects

Campagnoli, Carlo; Hung, Bethany; Domini, Fulvio

doi:10.1016/j.visres.2021.107961

Cited by 9 publications

(5 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we confirmed the prediction that increasing the number of cues specifying a 3D surface will increase the perceived depth of that surface, a hypothesis which we call the Vector Sum Model. This result has been recently found in other studies using grasping to test depth perception in both VR environments and with real objects (Campagnoli & Domini, 2019;2022). Although Bayesian models can account for the phenomenon predicted by the Vector Sum model, the IC theory has the significant advantage of achieving the same predictions without the need for further ad-hoc assumptions such as cues-to-flatness or priors-toflatness (Di Luca et al, 2010;Domini and Caudek, 2003, 2009, 2011, 2013Domini et al, 2006;.…”

Section: Discussionsupporting

confidence: 76%

Perceiving Depth from Texture and Disparity Cues: Evidence for a Non-Probabilistic Account of Cue Integration

Kemp

Cesanek

Domini

2022

Preprint

Self Cite

View full text Add to dashboard Cite

The fundamental question of how the brain derives 3D information from the inherently ambiguous visual input has been approached during the last two decades with probabilistic theories of 3D perception. Probabilistic models, such as the Maximum Likelihood Estimation (MLE) model, derive from multiple independent depth cues the most probable 3D interpretations. These estimates are then combined by weighing them according to their uncertainty to obtain the most accurate and least noisy estimate. In three experiments we tested an alternative theory of cue integration termed the Intrinsic Constraint (IC) theory. This theory postulates that the visual system does not derive the most probable interpretation of the visual input, but the most stable interpretation amid variations in viewing conditions. This goal is achieved with the Vector Sum model, that represents individual cue estimates as components of a multidimensional vector whose norm determines the combined output. In contrast with the MLE model, individual cue estimates are not accurate, but linearly related to distal 3D properties through a deterministic mapping. In Experiment 1, we measured the cue-specific biases that arise when viewing single-cue stimuli of various simulated depths and show that the Vector Sum model accurately predicts an increase in perceived depth when the same cues are presented together in a combined-cue stimulus. In Experiment 2, we show how Just Noticeable Differences (JNDs) are accounted for by the IC theory and demonstrate that the Vector Sum model predicts the classic finding of smaller JNDs for combined-cue versus single-cue stimuli. Most importantly, this prediction is made through a radical re-interpretation of the JND, a hallmark measure of stimulus discriminability previously thought to estimate perceptual uncertainty. In Experiment 3, we show that biases found in cue-integration experiments cannot be attributed to flatness cues, as assumed by the MLE model. Instead, we show that flatness cues produce no measurable difference in perceived depth for monocular (3A) or binocular viewing (3B), as predicted by the Vector Sum model.

show abstract

Section: Discussionsupporting

confidence: 76%

Perceiving Depth from Texture and Disparity Cues: Evidence for a Non-Probabilistic Account of Cue Integration

Kemp

Cesanek

Domini

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…First, we confirmed the prediction that increasing the number of cues specifying a 3D surface will increase the perceived depth of that surface, a hypothesis emerging from the vector sum model. This result has been recently found in other studies using grasping to test depth perception in both virtual reality environments and with real objects ( Campagnoli & Domini, 2019 ; Campagnoli, Hung, & Domini, 2022 ). Although Bayesian models can account for the phenomenon predicted by the vector sum model, the IC theory has the significant advantage of achieving the same predictions without the need for further ad hoc assumptions such as flatness cues or a flatness prior ( Di Luca, Domini, & Caudek, 2010 ; Domini and Caudek, 2003 ; Domini and Caudek, 2009 ; Domini and Caudek, 2010 ; Domini and Caudek, 2011 ; Domini and Caudek, 2013 ; Domini et al, 2006 ; Domini, Shah, & Caudek, 2011 ).…”

Section: Discussionsupporting

confidence: 84%

Perceiving depth from texture and disparity cues: Evidence for a non-probabilistic account of cue integration

2023

Self Cite

View full text Add to dashboard Cite

Bayesian inference theories have been extensively used to model how the brain derives three-dimensional (3D) information from ambiguous visual input. In particular, the maximum likelihood estimation (MLE) model combines estimates from multiple depth cues according to their relative reliability to produce the most probable 3D interpretation. Here, we tested an alternative theory of cue integration, termed the intrinsic constraint (IC) theory, which postulates that the visual system derives the most stable, not most probable, interpretation of the visual input amid variations in viewing conditions. The vector sum model provides a normative approach for achieving this goal where individual cue estimates are components of a multidimensional vector whose norm determines the combined estimate. Individual cue estimates are not accurate but related to distal 3D properties through a deterministic mapping. In three experiments, we show that the IC theory can more adeptly account for 3D cue integration than MLE models. In Experiment 1, we show systematic biases in the perception of depth from texture and depth from binocular disparity. Critically, we demonstrate that the vector sum model predicts an increase in perceived depth when these cues are combined. In Experiment 2, we illustrate the IC theory radical reinterpretation of the just noticeable difference (JND) and test the related vector sum model prediction of the classic finding of smaller JNDs for combined-cue versus single-cue stimuli. In Experiment 3, we confirm the vector sum prediction that biases found in cue integration experiments cannot be attributed to flatness cues, as the MLE model predicts.

show abstract

“…Instead, invariant properties of independent image signals that covary with distal properties must also necessarily covary among themselves and, likewise, covary with signals arising from other modalities such as touch or proprioception. Discovering these covariations may be a plausible learning process for a biological system that never has direct access to the distal geometric structure of the environment [ 85 , 86 ].…”

Section: Discussionmentioning

confidence: 99%

The case against probabilistic inference: a new deterministic theory of 3D visual processing

Domini

2022

Phil. Trans. R. Soc. B

Self Cite

View full text Add to dashboard Cite

How the brain derives 3D information from inherently ambiguous visual input remains the fundamental question of human vision. The past two decades of research have addressed this question as a problem of probabilistic inference, the dominant model being maximum-likelihood estimation (MLE). This model assumes that independent depth-cue modules derive noisy but statistically accurate estimates of 3D scene parameters that are combined through a weighted average. Cue weights are adjusted based on the system representation of each module's output variability. Here I demonstrate that the MLE model fails to account for important psychophysical findings and, importantly, misinterprets the just noticeable difference, a hallmark measure of stimulus discriminability, to be an estimate of perceptual uncertainty. I propose a new theory, termed Intrinsic Constraint, which postulates that the visual system does not derive the most probable interpretation of the visual input, but rather, the most stable interpretation amid variations in viewing conditions. This goal is achieved with the Vector Sum model, which represents individual cue estimates as components of a multi-dimensional vector whose norm determines the combined output. This model accounts for the psychophysical findings cited in support of MLE, while predicting existing and new findings that contradict the MLE model. This article is part of a discussion meeting issue ‘New approaches to 3D vision’.

show abstract

Explicit and implicit depth-cue integration: Evidence of systematic biases with real objects

Cited by 9 publications

References 38 publications

Perceiving Depth from Texture and Disparity Cues: Evidence for a Non-Probabilistic Account of Cue Integration

Perceiving Depth from Texture and Disparity Cues: Evidence for a Non-Probabilistic Account of Cue Integration

Perceiving depth from texture and disparity cues: Evidence for a non-probabilistic account of cue integration

The case against probabilistic inference: a new deterministic theory of 3D visual processing

Contact Info

Product

Resources

About